The overall goal of this proposal is to develop and exploit mouse models which faithfully recapitulate the biology and genetics of astrocytoma and oligodendroglioma, the most commonly occurring human primary brain tumors. These animals will be used to dissect the pathogenesis and biology of glioma and to improve efforts directed at cancer prevention, diagnosis, and therapy. A major additional emphasis is to contribute our unique resources and experience towards development of animal models for other human tumors studied by Consortium members. Gliomas are the leading cause of cancer deaths in children, and represent the fourth leading cause of cancer-related death in males aged 35-54. There is no current means to prevent gliomas, to diagnose gliomas before they become symptomatic, nor to effectively treat glioma. The incidence of gliomas is increasing in all age groups, and the highest grade tumors almost invariably causes death within a year of diagnosis. We propose to build upon current genetic insights to develop animal models which faithfully recapitulate the most common of these tumors, which are also among the most lethal. We have targeted expression of v-erbB, an oncogenic form of the epidermal growth factor receptor, to generate an infiltrating transgenic model for oligodendroglioma. We plan to characterize and validate these tumors, and to generate additional mice with tissue specific gain-of-function and loss-of-function mutations in genes comprising the main genetic pathways in human astrocytomas. We propose specific strategies for utilizing the mouse lines and tumors which arise from these experiments to improve our understanding of tumor biology, to identify additional genetic lesions, and to develop improved strategies for their diagnosis and treatment. Specifically we will: 1) Map and identify brain tumor modifier loci in the mouse, 2) Utilize state-of-the-art expression array technology to examine patterns of gene expression and develop array based comparative genomic hybridization in the mouse to identify new genes that contribute to the genetics of these tumors, and 3) Develop functional magnetic resonance imaging modalities for mouse brain tumors, and apply these technologies to both basic science and translational research questions. These experiments should improve our understanding of human glioma as well as glioma models, and assist in the development and evaluation of new therapeutic paradigms. In addition, the technologies developed through this proposal are readily applicable to animal models of other tumor types which might emerge from Consortium efforts.